Objective: Prior research suggests a bidirectional relationship between obesity and major depressive disorder (MDD), but the results have been heterogeneous. Differences between males and females in the association of MDD with obesity may contribute to inconsistent results. Thus, this study was designed to determine whether sex has a differential effect on the relationship between MDD and obesity, and to explore the potential mechanisms. Methods: All participants were diagnosed with MDD, and depression severity was measured using the 17-item Hamilton Depression Rating Scale. Body weight and height were measured to calculate body mass index (BMI). Body composition, including total fat, trunk fat, android fat, and visceral fat mass, was measured by dual-energy X-ray absorptiometry. Subjects provided blood samples, and serum was extracted for measuring the inflammatory factors using human immunoassay kits. Results: Among all obesity measures, depressed women had greater BMI and total body fat. By contrast, depressed men had greater visceral fat mass. However, only in depressed women was depression correlated with several measures of obesity, including BMI, total body fat, and visceral fat mass. A stepwise multiple regression analysis was conducted, and only visceral fat entered the regression model and was most predictive of depression in women (β = 0.60, p = 0.007). Moreover, compared with depressed men, depressed women had higher leptin levels after controlling for BMI, total body fat, and visceral fat. Conclusion: These results highlight gender differences in determining the association between obesity and depression, and elevated leptin level is a potential mechanism linking MDD to obesity in depressed women. Understanding a gender-specific relationship between obesity and MDD would allow clinicians to target and personalize therapies in the hope of improving health outcomes.